Development of a practical spatial-spectral analysis protocol for breast histopathology using Fourier transform infrared spectroscopic imaging

摘要: Breast cancer screening provides sensitive tumor identification, but low specificity implies that a vast majority of biopsies are not ultimately diagnosed as cancer. Automated techniques to evaluate can prevent errors, reduce pathologist workload and provide objective analysis. Fourier transform infrared (FT-IR) spectroscopic imaging both molecular signatures spatial information may be applicable for pathology. Here, we utilize the spectral develop combined classifier rapid tissue assessment. First, evaluated potential IR diagnosis using data alone. While highly accurate histologic [epithelium, stroma] recognition could achieved, same was possible disease [cancer, no-cancer] due diversity signals. Hence, employed data, developing evaluating increasingly complex models, detect cancers. Sub-mm tumors very confidently predicted indicated by quantitative measurement accuracy via receiver operating characteristic (ROC) curve analyses. The developed protocol validated with small set statistical performance used model predicts study design large scale, definitive validation. results evaluation on different instruments, at higher noise levels, under coarser resolution two sampling modes [transmission transflection], indicate is variety conditions. paves way validating breast detection, its validation directions optimization speed clinical deployment.